STEREO observations of HD90386 (RX Sex): a δ- Scuti or a hybrid star?

Transcription

1 Journal of Physics: Conference Series PAPER OPEN ACCESS STEREO observations of HD90386 (RX Sex): a δ- Scuti or a hybrid star? To cite this article: D Ozuyar et al 2016 J. Phys.: Conf. Ser View the article online for updates and enhancements. Related content - Erratum: Electromagnetic Waves near the Proton Cyclotron Frequency: STEREO Observations (2014, ApJ, 786, 123) L. K. Jian, H. Y. Wei, C. T. Russell et al. - Gradual Solar Coronal Dimming and Evolution of Coronal Mass Ejection in the Early Phase Jiong Qiu and Jianxia Cheng - EVIDENCE FOR HOT FAST FLOW ABOVE A SOLAR FLARE ARCADE S. Imada, K. Aoki, H. Hara et al. This content was downloaded from IP address on 30/04/2018 at 01:28

2 STEREO observations of HD (RX Sex): a δ-scuti or a hybrid star? D Ozuyar 1,2, I R Stevens 2, G Whittaker 3 and V Sangaralingam 4 1 Ankara University, Deparment of Astronomy and Space Sciences, Ankara, Turkey 2 The University of Birmingham, School of Physics and Astronomy, Birmingham, U.K. 3 The University of Toronto, Toronto Ontario, Canada 4 The University of Montreal, School of Physics, Montreal, Canada dozuyar@ankara.edu.tr, irs@star.sr.bham.ac.uk, whittaker@astro.utoronto.ca, vinothini.sangaralingam@asc-csa.gc.ca Abstract. HD is a rarely studied bright A2V type δ Scuti star (V = 6.66 mag). It displays short-term light curve variations which are originated due to either a beating phenomenon or a non-periodic variation. In this paper, we presented high-precision photometric data of HD taken by the STEREO satellite between 2007 and 2011 to shed light on its internal structure and evolution stage. From the frequency analysis of the four-year data, we detected that HD had at least six different frequencies between 1 and 15 c d 1. The most dominant frequencies were found at around c d 1 (A 1.92 mmag) and c d 1 (A 0.61 mmag). Based on the ratio between these frequencies, the star was considered as an overtone pulsator. The variation in pulsation period over 35 years was calculated to be dp/pdt = 5.39(2) x 10 3 yr 1. Other variabilities at around 1.0 c d 1 in the amplitude spectrum of HD90386 were also discussed. In order to explain these variabilities, possible rotational effects and γ Dor type variations were focused. Consequently, depending on the rotation velocity of HD 90386, we speculated that these changes might be related to γ Dor type high-order g-modes shifted to the higher frequencies and that HD might be a hybrid star. 1. Introduction δ Scuti stars are located at the intersection of the classical Cepheid instability strip and the main sequence (MS). Their luminosity classes vary between V to III, since some massive samples evolve towards the giant region and across the strip at higher luminosities. Spectral types of these stars range from A2 to F0 on the MS and from A3 to F5 in the giant region [1]. Also, the mass values are between 1.5 and 2.5 M for δ Scuties with solar metal abundance and M for metal-poor samples. Therefore, it is known that these stars are either at the core or shell hydrogen-burning stages. They are mostly non-radial pulsators, but some δ Scuties show radial oscillations. The periods of the pulsations are between 0.02 days and 0.25 days, with a typical amplitude of around 0.02 mag [2]. HD is a rarely studied A2V type, bright δ Scuti star (B = 6.78 mag, V = 6.66 mag). Its variability was accidentally discovered by Jerzykiewicz [3] while it was being used as a comparison star for 23 Sex. He detected short-term variations with the period of days (A 0.01 mag) in the light curve (LC), and reported that their origins were due to either a beating phenomenon or a non-periodic variation in LC. Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. Published under licence by Ltd 1

3 2. STEREO Satellite and Data Analysis In this paper, we present high-precision photometric data of HD taken by the STEREO satellite to shed light on its internal structure and evolution stage. During the seasonal observations, 684, 678, 651, and 660 data points were derived from the satellite in 2007, 2008, 2010, and 2011, respectively. This yielded a total of 2,673 points, corresponding to 1,809 hours of observation over four years. The data were provided only from the HI-1A instrument on-board the STEREO-A satellite. It has been capable of observing background stars with the magnitude of 12 m or brighter for a maximum of 20 days and a useful stellar photometer which has covered region around the ecliptic (20% of the sky) with the field of view (FOV) of A more detailed description of the data preparation and background information on STEREO can be found in Sangaralingam and Stevens [4] and Whittaker et al. [5]. The nominal exposure time of the camera is 40 seconds, and putting 30 exposures together on board, a 40-minute integrated cadence has been obtained to transmit for each HI-1 image (For details of the HI instruments refer to Eyles et al. [6]). Therefore, the Nyquist frequency of the data is around 18 c d 1. Since the STEREO satellite has observed the region quite close to the solar disk, data of HD were mostly affected by solar activities. Therefore, the distorted effects were removed from the LCs by using a 3 rd order polynomial fit, and observation points greater than 3σ were clipped with the help of a freshly coded pipeline in IDL. Four years of seasonal and combined time series were then analysed with the Lomb-Scargle (LS) algorithm. The results found were compared to those of Period04. For each periodogram, regional noise levels were determined by averaging the noise values in every 0.5 c d 1, and a specific noise characteristic was thus established. Based on this characteristic, a significance level was calculated with 99% probability. Frequencies whose amplitudes were greater than this level were then detected. The detection precision in four-year combined data given in Figure 1 was around 10 5 c d 1 in frequency and 10 4 mag in amplitude. Figure 1: The amplitude spectrum of HD90386 obtained from four-year combined light curves. The mean noise level is plotted with red dash-dot line. Red dashed line is the significance level calculated based on the noise characteristics, and blue dashed line is the 4σ level. The estimated rotational periods are shown with green and red vertical lines. For more details, see the text. 2

4 Table 1: The frequencies of HD 90386, their amplitudes, SNR values, noise levels (A m ), and pulsation constants (Q). HD Frequencies No Freq. Amp. SNR A m Q (c d 1 ) (mmag) (mmag) (days) f (2) 1.92(10) (6) f (5) 0.74(10) f (6) 0.61(10) (5) f (8) 0.50(10) f (11) 0.36(10) (8) f (11) 0.34(10) (6) 2007 f (2) 1.94(16) f (6) 0.77(16) f (8) 0.59(16) f (3) 1.98(17) f (8) 0.65(17) f (3) 1.94(20) f (1) 0.55(20) f (4) 1.77(22) f (10) 0.69(22) f (12) 0.55(22) The noise and related significance levels are presented with red dash-dot and red dashed lines in the figure, respectively. For a comparison, the constant significance level calculated from the mean noise value of the periodogram is shown with a blue dashed line. Additionally, the extracted frequencies, their amplitudes, signal-to-noise ratios (SNR) and noise (A m ) values are given in Table Results and Discussion The frequency analysis of the combined data set showed that there were six frequencies having amplitude greater than the specific significance level in the amplitude spectrum (Table 1). As seen in Figure 1, four of these peaks were located between the frequencies of 5 and 15 c d 1. Among them, the most dominant frequency was found at around c d 1 ( days) with the amplitude of 1.92 mmag. Also, the second powerful frequency showed up at c d 1 ( days) in our amplitude spectrum, and it was quite close to the result ( days) obtained by Jerzykiewicz [3]. Based on this alteration in the main frequency, it might be suggested that the star gradually switched pulsation modes from to c d 1 over time. Furthermore, the intensity of the c d 1 in the seasonal LCs seemed to decrease between 2007 and As a result of this decrement, its amplitude observed in 2010 and 2011 fell below the significance level (Figure 2). However, their SNR values determined from Period04 were over 4.0, and they were hence included in the analyses. 3

5 Figure 2: Seasonal light curves of HD and their amplitude spectra. As seen in Figure 2, all seasonal LCs exhibit variabilities around the region of 1.00 c d 1. Since there was not a sufficient number of studies on the physical, chemical, or orbital parameters of the star, interpretation of these variations might not be possible. However, Niemczura et al. [7] demonstrated that the projected rotational velocities of typical A and F stars ranged from 8 to about 280 km s 1, with a mean of 134 km s 1 in the analysis of high-resolution spectroscopic data of A/F type stars from the Kepler field. Moreover, they found that the maximum of the velocity distribution for A and F stars having an effective temperature between 8,000 and 9,000 K was around 200 km s 1. Assuming that the temperature and radius of the star were around logt = ( 8200 K; approximated from the B V colour index) and R = 2.85 R (calculated from the luminosity equation L/L = [R/R ] 2 [T /T ] 4 ), its rotational period was 4

6 estimated to be 0.93(7) and 1.39(11) c d 1 for an inclination angle of i = 90, and rotational velocities of 134 and 200 km s 1, respectively. These values shown with green and red vertical lines in Figure 1 were considerably compatible with the frequency accumulation at around 1.00 c d 1. We also computed the pulsation constants (Q) of the δ Scuti type frequencies by using the relation provided by Breger and Bregman [8]. Consequently, Q values of f 1 and f 6 were around 0.031(6) and 0.034(6) days, which were within the theoretical range for the fundamental mode given by Fitch [9]. If the main frequency was the fundamental mode, its first overtone would be around c d 1, which was not observed in our amplitude spectrum. In contrast, if 9.37 c d 1 was assumed to be the fundamental mode, the first overtone was found as c d 1 based on the period ratio P 1 /P 0 = This value slightly matched with f 3, and its Q constant was also within the theoretical range given for the first overtone. Additionally, seasonal frequency and amplitude variations were investigated to gain better understanding of the short-term pulsation behaviour of RX Sex. To do this, we determined consistently observed frequencies in seasonal LCs (Figure 2) and surmised that only two frequencies potentially varied as seen in Figure 3. However, all variations were within the error limits, and no significant frequency or amplitude changes were found for both peaks over four years. On the other hand, combining the frequency value ( c d 1 ) given by Jerzykiewicz [3] with our STEREO result, a long-term period variation of 2.49(22) 10 4 yr 1 was calculated for the frequency at around c d 1. It should also be noted that if there was a gradual change between the frequencies (from to c d 1 ), the variation rate of the period would be obtained as dp/p dt = 5.39(2) 10 3 yr 1 over 35 years. According to Breger and Pamyatnykh [10], this suggested that the period change observed was not caused by stellar evolutionary changes, but produced by another factor such as non-linear mode interactions. Figure 3: Two frequencies of HD and their amplitude variations between 2007 and

7 In order to have a better understanding of the evolutionary stage of HD 90386, we attempted to estimate some of the fundamental physical parameters. Its effective temperature and luminosity were approximated to be around logt = 3.914(4) from the B V colour index and logl/l = 1.517(67) from the parallax given in Simbad Database, respectively. Using these parameters, the radius of the star was derived as 2.85(23) R. The surface gravity (logg = 3.887(159)) was obtained from the relation given by Tsvetkov [11] and the mass (M/M = 2.28(9)) was computed by means of the equation taken from Cox and Pilachowski [12]. The luminosity and temperature values indicate that HD is quite close to the blue edge of the instability strip and to the end of the MS stage where central hydrogen is consumed (Figure 4). Its metallicity index (m 1 = mag) is low [13]. So, this might result in higher luminosity and temperature values, which specify its position on the H-R diagram. Its mass value is also compatible with the theoretical evolution track. Further, since the inclination angle is not known and the rotational velocity is ambiguous, the effects of the rotation on the amplitude spectrum and the evolution status are not clear. Figure 4: The evolutionary stage of HD The blue and red lines are the hot and cold edges of the instability strip, respectively. The vertical dashed line is the hot edge for the fundamental mode pulsators (Breger & Pamyatnykh, 1998b). The theoretical values in the MS and the post-ms phases are taken from Christensen-Dalsgaard [14]. HD 90386, a double-mode pulsator, has a simple amplitude spectrum. The period ratio between f 1 and f 3 is around 0.827, which is quite consistent with the ratio given by Breger [2] for the first and second overtones (P 2 /P 1 = 0.810). Based on Breger and Bregman [8], HD might be an overtone pulsator since δ Scuties pulsating with the fundamental mode lie in the cool part of the instability strip whereas the overtone pulsators are found in the hotter part. As previously mentioned, the variabilities around 1.00 c d 1 in the amplitude spectrum are consistent with the rotation period. However, there is a possibility that these variations are related to γ Dor type high-order g-modes and that HD might be a δ Scuti/γ Dor type hybrid star. Although the location of the star on the H-R diagram suggests that it is a δ Scuti 6

8 type star, in the studies related Kepler A and F type stars Bradley et al. [15] and Uytterhoeven et al. [16] showed that the locations of δ Scuti and γ Dor classes on the H-R diagram have been extended, and that γ Dor type pulsations have been seen in hotter stars than those known so far. They observed a great number of hybrid stars which spread between the blue edge of the δ Scuti instability strip and the red edge of the γ Dor instability strip (two-third of the hybrids have been located in the hot side of the δ Scuti instability strip [15]. In this context, the frequency accumulation at around 1.00 c d 1 might in fact be related to γ Dor type variations and HD might be thought as a hybrid star based on the results of these studies. Bear in mind that the stellar rotation might also affect the frequency distribution of the star. According to Bouabid et al. [17], if the star is a fast rotator, γ Dor type pulsations have been shifted to higher frequencies and filled the gap between γ Dor type high-order g-modes and δ Scuti type modes, which is compatible with the frequency range of 5 and 15 c d 1. On this basis, such a situation might explain the low-amplitude variations seen between 5 and 15 c d 1 in the periodogram of the star. Even though HD is classified as a δ Scuti type pulsator in the literature, it is quite difficult to conclude the actual class of the star by using only the STEREO data. Therefore, longterm follow-ups similar to those of Kepler satellite, multi-colour photometry, and spectroscopic observations are needed in order to accurately determine physical parameters and the nature of the star. References [1] Kurtz D W 2000 ASPC [2] Breger M 1979 PASP 91 5 [3] Jerzykiewicz M 1972 PASP [4] Sangaralingam V and Stevens I R 2011 MNRAS [5] Whittaker G N, Stevens I R and Sangaralingam V 2013 MNRAS 431, 3456 [6] Eyles C J, et al SoPh [7] Niemczura E, et al MNRAS [8] Breger M and Bregman J N 1975 ApJ [9] Fitch W S 1981 ApJ [10] Breger M and Pamyatnykh A A 1998 A&A [11] Tsvetkov T G 1988 Ap&SS [12] Cox A N and Pilachowski C A 2000 PhT [13] Rodríguez E, López-González M J and López de Coca P 2000 A&AS [14] Christensen-Dalsgaard J 1993 ASPC [15] Bradley P A, Guzik J A, Miles L F, Uytterhoeven K, Jackiewicz J and Kinemuchi K 2015 AJ [16] Uytterhoeven K, et al A&A 534 A125 [17] Bouabid M P, Dupret M A, Salmon S, Montalbán J, Miglio A and Noels A 2013 MNRAS